This file is part of IDEAS, which uses RePEc data


[ Papers | Articles | Software | Books | Chapters | Authors | Institutions | JEL Classification | NEP reports | Search | New papers by email | Author registration | Rankings | Volunteers | FAQ | Blog | Help! ]

Bimodal t-Ratios

Author info | Abstract | Publisher info | Download info | Related research | Statistics
Author Info
Peter C.B. Phillips () (Cowles Foundation, Yale University)
Vassilis A. Hajivassiliou (Cowles Foundation, Yale University)

Additional information is available for the following registered author(s):

Abstract

This paper studies the sampling distribution of the conventional t-ratio when the sample comprises independent draws from a standard Cauchy (0,1) population. It is shown that this distribution displays a striking bimodality for all sample sizes and that the bimodality persists asymptotically. An asymptotic theory is developed in terms of bivariate stable variates and the bimodality is explained by the statistical dependence between the numerator and denominator statistics of the t-ratio. This dependence also persists asymptotically. These results are in contrast to the classical t statistic constructed from a normal population, for which the numerator and denominator statistics are independent and the denominator, when suitably scaled, is a constant asymptotically. Our results are also in contrast to those that are known to apply for multivariate spherical populations. In particular, data from an n dimensional Cauchy population are well known to lead to a t-ratio statistic whose distribution is classical t with n-1 degrees of freedom. In this case the univariate marginals of the population are all standard Cauchy (0,1) but the sample data involves a special form of dependence associated with the multivariate spherical assumption. Our results therefore serve to highlight the effects of the dependence in component variates that is induced by a multivariate spherical population. Some extensions to symmetric stable populations with exponent parameter alpha does not equal 1 are also indicated. Simulation results suggest that the sampling distributions are well approximated by the asymptotic theory even for samples as small as n = 20.

Download Info
To download:

If you experience problems downloading a file, check if you have the proper application to view it first. Information about this may be contained in the File-Format links below. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.

File URL: http://cowles.econ.yale.edu/P/cd/d08a/d0842.pdf
File Format: application/pdf
File Function:
Download Restriction: no

Publisher Info
Paper provided by Cowles Foundation, Yale University in its series Cowles Foundation Discussion Papers with number 842.

Download reference. The following formats are available: HTML (with abstract), plain text (with abstract), BibTeX, RIS (EndNote, RefMan, ProCite), ReDIF
Length: 26 pages
Date of creation: Jul 1987
Date of revision:
Handle: RePEc:cwl:cwldpp:842

Contact details of provider:
Postal: Yale University, Box 208281, New Haven, CT 06520-8281 USA
Phone: (203) 432-3702
Fax: (203) 432-6167
Web page: http://cowles.econ.yale.edu/
More information through EDIRC

Order Information:
Postal: Cowles Foundation, Yale University, Box 208281, New Haven, CT 06520-8281 USA

For technical questions regarding this item, or to correct its listing, contact: (Glena Ames).

Related research
Keywords:

Other versions of this item:

References listed on IDEAS
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
  1. Peter C.B. Phillips, 1982. "Exact Small Sample Theory in the Simultaneous Equations Model," Cowles Foundation Discussion Papers 621, Cowles Foundation, Yale University. [Downloadable!]
    Other versions:
Full references

Cited by:
(explanations, Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.)

  1. Peter C.B. Phillips, 1987. "Conditional and Unconditional Statistical Independence," Cowles Foundation Discussion Papers 824R, Cowles Foundation, Yale University, revised Dec 1987. [Downloadable!]
    Other versions:
  2. Peter C.B. Phillips & Mico Loretan, 1990. "Testing Covariance Stationarity Under Moment Condition Failure with an Application to Common Stock Returns," Cowles Foundation Discussion Papers 947, Cowles Foundation, Yale University. [Downloadable!]
  3. Peter C.B. Phillips & Mico Loretan, 1989. "The Durbin-Watson Ratio Under Infinite Variance Errors," Cowles Foundation Discussion Papers 898R, Cowles Foundation, Yale University, revised Aug 1989. [Downloadable!]
    Other versions:
  4. Peter C.B. Phillips, 1989. "Time Series Regression with a Unit Root and Infinite Variance Errors," Cowles Foundation Discussion Papers 897R, Cowles Foundation, Yale University, revised Aug 1989. [Downloadable!]
    Other versions:
Statistics
Access and download statistics

Did you know? A few items listed on IDEAS are over 2000 years old!

This page was last updated on 2009-11-12.


This information is provided to you by IDEAS at the Department of Economics, College of Liberal Arts and Sciences, University of Connecticut using RePEc data on a server sponsored by the Society for Economic Dynamics.